Functions to design and apply tests that are anytime valid. The
functions can be used to design hypothesis tests in the prospective/randomised
control trial setting or in the observational/retrospective setting. The
resulting tests remain valid under both optional stopping and optional
continuation. The current version includes safe t-tests and safe tests of
two proportions. For details on the theory of safe tests, see Ly, A, Boehm,
Grunwald, Ramdas and van Ravenzwaaij (2024). "Safe Anytime-Valid Inference:
Practical maximally flexible sampling designs for experiments based on e-values"
(
Safestats is for designing and applying safe hypothesis tests. It can be used for designing hypothesis tests in the prospective or randomised controlled trial (RCT) setting, where the tests can be used under optional stopping allowing experiments to be stopped early without the loss of statistical validity. The 'pilot' functions in the package also enable using the safe tests in the observational/ retrospective setting. For examples and explanation about which test to choose for which study setup, our vignette can be used (see below). The current version includes safe t-tests and tests of two proportions. The initial paper on the theory of safe testing and a worked-out example for the t-test can be found in this paper. More on the theory behind the development of the safe tests for proportions can be found here.
The development version can be found on GitHub, which can be installed with the devtools package from CRAN by entering in R:
devtools::install_github("AlexanderLyNL/safestats", build_vignettes = TRUE)
Once a stable version is finished, you can install it from CRAN in R through:
install.packages("safestats")
Load the safestats package:
library("safestats")
For an extensive overview of all functions and their usage, we recommend following the safestats vignette tutorial:
browseVignettes("safestats")